BUS504: Module 2 out of 6 begins.

I figured I should break up my assignments a little better so I don't spend my entire weekend doing homework again... so unless I am too tired, I should try to stay on schedule.  However, not doing any homework today since I need a break.

Discussion forum question [I will answer this sometime either Tuesday or Wednesday during lunch.]
Research questions and hypothesis-testing In the readings, a distinction is made between the general "research questions" that shape a study and the specific hypotheses it purports to test. So here's the question -- are there meaningful research questions (emphasis on "meaningful") that cannot be assessed through the process of operationalizing and testing hypotheses? If you believe that there are such questions, what might be an example of one (or more)? If you believe that there aren't, why not?

Dive in on this one -- you aren't being assessed on the degree to which you agree with my position on this question (which you don't actually know), but on the degree to which you can make an informed critical statement about it - and engage the statements of your colleagues. Maybe consensus will emerge, maybe not. It will be entertaining to see.

As usual, creative tearing into each other -- all in a spirit of good collegial fun, of course -- is encouraged strongly.



Case Assignment

Part One: [I will read during the week at lunch time]

In this module we'll continue to practice learning research by looking into actual research projects. Both Module 2 and Module 3 will make extensive use of this recent article:

Bacharach, S. & Bamberger, P. (2007). 9/11 and New York City firefighters' post hoc  unit support and control climates:  A context theory of the consequences of involvement in traumatic work-related events. The Academy of Management Journal, 50 (4): 849-868.

It's a study of New York City firemen in the aftermath of the events of 9/11.  It's interesting and important in its own right, and also provides excellent examples of variable operationalization and hypothesis formulation.  You should read the article down to the "Results" section, where the statistical analysis begins.  Then, skip to "Discussion," and read the article to the end.  You may need to do this several times.

Tip:  As you read, try to translate the academic jargon into ordinary English.  Example:  "Hypothesis 1.The association between the intensity of employees' involvement in a work-related critical incident and their post-event negative emotional states (i.e., depression, anxiety and stress) varies significantly over work units."  Translation:  People involved in disasters tend to suffer emotional injury.  Those who are more closely involved with the disaster suffer greater injury.  But the extent of the injury caused by a particular level of involvement depends on the characteristics of the work unit they're in. It helps to keep the context --i.e., the 9/11 attacks -- in mind as you read.

Required Readings

Bacharach, S. & Bamberger, P. (2007). 9/11 and New York City firefighters' post hoc  unit support and control climates:  A context theory of the consequences of involvement in traumatic work-related events. The Academy of Management Journal, 50 (4): 849-868. Retrieved May 25, 2009, from http://web.ebscohost.com.lb-proxy6.touro.edu/ehost/viewarticle?dat a=dGJyMPPp44rp2%2fdV0%2bnjisfk5Ie46bFOsqi1Srak63nn5Kx95uXxjL6rrUq0pbBIr q%2beSbimtVKzrp5oy5zyit%2fk8Xnh6ueH7N%2fiVaunsk2zqK9Lr6%2b1PurX7H%2b72% 2bw%2b4ti7evLepIzf3btZzJzfhruotEq1r69Qrpzkh%2fDj34y75uJ%2bxOvqhNLb9owA& hid=105  

Figley, C. (2008).  Impact of Event Scale Revised.  Retrieved on Feb 20, 2009 from http://www.atft.org/research/Impact%20of%20Event%20Scale%20-%20Revised.htm 

University of New South Wales (UNSW)  (n.d. a)  Overview of the DASS and its uses.  Retrieved on 25 Feb 2008 from http://www2.psy.unsw.edu.au/groups/dass//over.htm

University of New South Wales (UNSW) (n.d. b).  DASS21 survey form.  Retrieved on 25 Feb 2008 from http://www2.psy.unsw.edu.au/groups/dass//Down_W6_US/dass21.doc

Optional Readings

Dallal, G. (2004) Units of Analysis. The Little Handbook of Statistical Practice.  Retrieved May 25, 2009, from http://www.jerrydallal.com/LHSP/units.htm

Dereshiwsky, M.  (1998) Understanding Variables. Introduction to Research. Retrieved May 25, 2009, from http://jan.ucc.nau.edu/~mid/edr610/class/variables/variables/lesson3-1-1.html

Eveland, JD (2003) Some observations on data, variables, and relationships.  PowerPoint presentation.  Available here.

Trochim, W. (n.d.). Measurement. Social Research Methods Database. Retrieved May 25, 2009, from http://www.socialresearchmethods.net/kb/measure.htm

Part Two: [I will complete on Sunday]

You are to develop a brief critique of the Bacharach et al. study, in accordance with the following outline:

1. In one or two paragraphs, summarize the study, its findings, and its implications for management practice. [If you'd like a little more structure to this part of the exercise, click here.]
   
2. Identify the variables in the study.  Characterize them as DVs, IVs, MVs, or EVs.  State the level of analysis of each; either individual level, or group level.  Summarize your analysis in the form of a table assigning a code to each variable, including the type of the variable (see Introduction), the level of analysis (either I for individual, or G for group), and the full description of the variable. Here's such a table, with two lines filled in:
   

Code	Type	Level	Authors' label
Str	DV	I	Stress
Icv	IV	I	Intensity of critical incident involvement
(etc.)

3. Next, summarize each hypothesis using the variable codes.  One of these would be (from Figure 1):
   
Icv is positively correlated with Str.
Some of the hypotheses involve either moderation or mediation.  In that event, use the formula:
X is (positively / negatively) correlated with Y, (moderated / mediated) by Z
[What's the point? Is this just busy work? Absolutely not.  Both scientists and managers, when they are reading and studying a paper that they absolutely MUST master, do exactly this sort of thing.  Reading a research report is a labor-intensive activity!]
4. 
   
   Once you've completed your description of the variables and the listing of the hypotheses, conclude with your best estimate of the overall quality of this study as research, on the basis of what you know now.  Trust your judgment.  Be sure to explain the reasoning behind your assessment.

SLP [I will do the following Sunday]
For the Module 2 component of the Project, please review the feedback you received on Module 1, and make necessary revisions.  Then add the following.

• Characterize your variables in terms of (1) level of measurement (2) level of analysis, and (3) whether each of them is a DV, IV, MV, of EV.  Don't worry about operationalizing the variables at this point.  That's a topic for Module 3.
  
• State your research questions in terms of testable hypotheses.  As mentioned above, the preferred format is as follows:  "(DV) is (positively / negatively) correlated with (IV)."  There must be a separate hypothesis for each combination of DV and IV.  JUSTIFY each hypothesis, on the basis of what you've read and/or observed, either in this or in some other organization.
  

The second bullet above deserves additional explanation, and an extended example.  Suppose the DVs are job satisfaction and intention to quit, while the IVs are job tenure (time on the job) and pay.  You think that tenure and pay are predictors of satisfaction, but only tenure is a predictor of intention to quit.  Further, you think that higher pay and longer tenure are related to higher levels of job satisfaction, and that longer tenure is related to a decreased intention to quit.  Your hypotheses would then be:

• H1:  Pay is positively correlated with job satisfaction.
• H2:  Tenure is positively correlated with job satisfaction.
• H3:  Tenure is negatively correlated with intention to quit.

Be sure to provide precise and detailed arguments in support of each of your hypotheses.  Avoid statements such as, "it is commonly known that" or "it is obvious that."  Few things are commonly known, and even fewer things are obvious.  Whenever possible, cite concrete examples supporting your hypothesis.  Example:  If Outward Bound training increased productivity in companies A, B, and C, it is reasonable to hypothesize that it will have the same effect in your company (assuming that A, B, C, and your company are all reasonably similar in terms of structure, culture, and other key variables.)